Frederick Stein

I am back from holidays. The feature tested here is similar to the `GROUP_SIZE` feature in the `WF_CORRELATION` section tested in `QS/regtest-mp2-block`. Still, it is strange that the tests in...

Thank you for the info. This flag is related to the offloading of FFTs. This part needs some revamp anyways because of the poor performance of this kernel on GPUs...

As far as I am aware, NMR calculations on the MP2 level are not implemented. They will probably also not work out of the box as the respective code paths...

Furthermore, the derivatives of the MP2 energy within the direct canonical approach have not been implemented yet. So, there is no way to perform the requested calculation.

That should work. But I will not be able to help you with setting these kind of calculations up as I have never dealt with NMR calculations myself.

I have to correct myself. As it turns out (see [here](https://groups.google.com/g/cp2k/c/BXUnzjL0rcg/m/aSKWa6fPAQAJ)), NMR shift calculations are NOT supported with hybrid functionals.

Dear @shyking617 , Usually, there are no adjustments of the input files necessary. My first guess are grid operations considering your input files but I need more information to confirm...

Regarding your second issue: I have checked the routines raised errors/warnings in your output file. CP2K does not make use of them directly suggesting that they are used by routines...

Thank you for your report. I am currently attempting to fix it for the regular build.

With the ordinary build system, I can build and compile CP2K with LibXC 7.0.0 (AMD hardware).